The Directed Prediction Change - Efficient and Trustworthy Fidelity Assessment for Local Feature Attribution Methods
This work addresses the need for efficient and trustworthy explanation methods in high-stakes domains like medicine, though it is incremental as it builds on the existing Prediction Change metric.
The paper tackled the problem of evaluating the fidelity of local feature attribution methods in machine learning, proposing a new metric called Directed Prediction Change (DPC) that achieves an almost tenfold speedup and eliminates randomness compared to existing methods like Infidelity.
The utility of an explanation method critically depends on its fidelity to the underlying machine learning model. Especially in high-stakes medical settings, clinicians and regulators require explanations that faithfully reflect the model's decision process. Existing fidelity metrics such as Infidelity rely on Monte Carlo approximation, which demands numerous model evaluations and introduces uncertainty due to random sampling. This work proposes a novel metric for evaluating the fidelity of local feature attribution methods by modifying the existing Prediction Change (PC) metric within the Guided Perturbation Experiment. By incorporating the direction of both perturbation and attribution, the proposed Directed Prediction Change (DPC) metric achieves an almost tenfold speedup and eliminates randomness, resulting in a deterministic and trustworthy evaluation procedure that measures the same property as local Infidelity. DPC is evaluated on two datasets (skin lesion images and financial tabular data), two black-box models, seven explanation algorithms, and a wide range of hyperparameters. Across $4\,744$ distinct explanations, the results demonstrate that DPC, together with PC, enables a holistic and computationally efficient evaluation of both baseline-oriented and local feature attribution methods, while providing deterministic and reproducible outcomes.